检索结果-南通市图书馆

第25届中国控制与决策会议

作者： Yuxia Sun Cheng Liu hongkun qiu Engineering Training Center Shenyang Aerospace University

ISBN: (纸本)9781467355322

NoGo game is similar to Go game, that means they have similar notion of group, killing, stones and chessboard. However, their rulers are completely different, or opposite, like the first player suicides, kills a group, or has no more legal moves, the opponent will win the game. In this paper, the patterns of NoGo are expounded in detail. The pattern values, the pattern sizes and the pattern formats are discussed. At the same time, an optimizing pattern matching algorithm is put forward to find out the best move for NoGo game. In addition, if the best move has not been found by the matching pattern algorithm, a modified UCT algorithm is also applied to play the NoGo game. The experimental results show that the proposed approaches are workable for NoGo.

关键词： NoGo Computer Games Patterns and UCT algorithm

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Research on Techniques of Computer Game

Research on Techniques of Computer Game

引用

第25届中国控制与决策会议

作者： hongkun qiu Fuxiang Gao Yajie Wang Yan Ge Quan qiu Huimin Sun Engineering Training Center Shenyang Aerospace University Institute of Information Science and Engineering Northeastern University School of Information Science and Engineering Shenyang ligong University

ISBN: (纸本)9781467355339

Computer game is regarded as a new research field of Artificial Intelligence. It’s very difficult to design and implement a high-level computer game system. Some techniques of computer game were presented with an example from a chess game-Connect-six. Firstly, introduced the game tree and searching techniques. Secondly, the key technologies for computer games were discussed, including establishment and application of Hash Table, improvement the efficiency of pruning. Finally, some contrasting experiments were performed to test the performance of the program which used the previously mentioned techniques. Z he experiments’ results indicated that these techniques are very useful for computer game. And they have been confirmed by a series of competitions and tests. By these researches, a high-level game system can be produced and the results and conclusions of this paper also have some reference values to other computer game procedures.

关键词： Computer Game Minimax Algorithm Alpha-Beta Pruning Hash Table Evaluation

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Implementation of Common Platform for Two-Person Game

Implementation of Common Platform for Two-Person Game

引用

第25届中国控制与决策会议

作者： hongkun qiu Fuxiang Gao Yajie Wang Xuxu Song Quan qiu Yuxia Sun Engineering Training Center Shenyang Aerospace University Institute of Information Science and Engineering Northeastern University School of Information Science and Engineering Shenyang ligong University

ISBN: (纸本)9781467355339

Computer Games（CG） is considered to be one of the most challenging research directions in the field of artificial intelligence. Though it is one of the most active research areas in recent years, there are less researches about the platform for CG. So some technologies about common platform for two-person game are researched. According to the characteristics of computer game, such as Dots and Boxes, Surakarta, Connect-six and Amazons, an application example was designed by using C/C++ language. The main process were listed as follow: Firstly of all, specified concepts and routines for interface, then defined some important data, included data structures, object classes, variables and functions Secondly introduced the expression of chess board and pieces in the computer. And then created input and output channel for searching engine. So a common platform of two-person player game was implemented. Finally, a series of experimentation were done. And software testing are added in the software development process. The experimentation results proved that this platform is effective for computer game and contributed to reduce subsequent development cost.

关键词： Computer Game Common Platform Search Engine User Interface Input and Output Channel

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Solve Game of the Amazons with Neural Networks and Distributed Computing

Solve Game of the Amazons with Neural Networks and Distribut...

引用

第30届中国控制与决策会议

作者： Yuanxing Chang hongkun qiu School of Computer Science Shenyang Aerospace University Engineering Training Center Shenyang Aerospace University

A training model was constructed with convolution neural network. Its policy network was trained by supervised learning. Then the parameters of the policy network were constantly adjusted by self-play and the game strength was enhanced. So it made it surpassed the existing game programs that simply used-pruning algorithm or Monte Carlo algorithm. By using distributed computing to update the data asynchronously, multiple game groups were allocated to different machines at the same time. In this way, a number of common configuration computers could be made full use of.

关键词： Game of the Amazons Neural Networks Distributed Computing

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Intelligent Design of Computer Bridge

Intelligent Design of Computer Bridge

引用

第32届中国控制与决策会议

作者： Jize Wang hongkun qiu Yuxi Chen Yajie Wang College of Computer Science Shenyang Aerospace University Engineering Training Center Shenyang Aerospace University School of Aero-engine Shenyang Aerospace University

Computer games are divided into complete information games and incomplete information games. Bridge is a type of incomplete information games. Due to the limited information that each player knows, the existing methods for solving complete information games cannot be directly applied to the bridge. Therefore, improving the ability of computer bridge agents is of great significance to the research in the field of artificial intelligence. This paper proposes a new bridge game system design structure and applies an improved Monte Carlo sampling algorithm to the computer bridge. Aiming at the problem that the state space information in the incomplete information game is huge and the optimal card can not be found in a reasonable time, a filtering mechanism is proposed to filter the sampling. And, a bidding module is designed, which can handle the stages of response and call when bidding. The experiment results show that the performance of the computer bridge system in this paper has certain reference value for incomplete information games.

关键词： Bridge Computer Games Incomplete Infomation Monte Carlo

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Analysis of Search Algorithm in Computer Game of Amazons

Analysis of Search Algorithm in Computer Game of Amazons

引用

第26届中国控制与决策会议

作者： qiu hongkun Zhang Peng Wang Yajie Wu Jiehong Li Fei Engineering Training Center Shenyang Aerospace University School of Aerospace Engineering Shenyang Aerospace University

ISBN: (纸本)9781479937097

Computer game is a new research field. This paper described rules of the Amazons and discussed some popular searching technologies in computer game of Amazons, such as the PVS algorithm, MTD-f algorithm and MTD-bi(bisection) algorithm. Then a series of experiments were done to measure their effectiveness and efficiency. The experimental results of performance testing revealed that sorting orders can increase performance of each algorithm in computer game of amazons, but PVS gets faster and performs better comparatively. And it had been verified in the National Computer Games Tournament.

关键词： Computer Game PVS MTD-f MTD-bi

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Computer-Bridge Software Test Platform Based on Anonymous Pipe

Computer-Bridge Software Test Platform Based on Anonymous Pi...

引用

第32届中国控制与决策会议

作者： Chen Yuxi qiu hongkun Wang Jize Ni Zijian Wang Yajie School of Aero-engine Shenyang Aerospace University Engineering Training Center Shenyang Aerospace University College of Computer Science Shenyang Aerospace University

The significant momentum of computer-bridge brings a strong demand for quality validation and assurance for computer-bridge software. Current research work seldom discusses corresponding software testing questions and validation approaches with clear quality requirements and methods. This paper focuses on a platform based on anonymous pipes that can load different game engines for multi-program battles or man-program battles, including the essential problems in the platform, features, and process. Moreover, it presents a test process for a certain computer-software. Finally, it discusses the challenges, issues, and needs in computer-software testing.

关键词： Computer Bridge Software Testing Game Platform Contract Bridge

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

引入合作竞争关系的桥牌叫牌数据库构建

引用

重庆理工大学学报（自然科学） 2022年第12期36卷 142-147页

作者：邱虹坤郑晓东王亚杰沈阳航空航天大学计算机学院沈阳110136 沈阳航空航天大学工程训练中心沈阳110136

传统的叫牌数据大多是对于自然叫牌法或精确叫牌法的存储,着重考虑扩充规则,较少考虑到牌手与牌手之间的相互配合。结合数据库智能化的思想,构建了增加竞争与合作标识的叫牌数据库。使其能够描述桥牌比赛中存在的合作竞争关系。此外,通... 详细信息

传统的叫牌数据大多是对于自然叫牌法或精确叫牌法的存储,着重考虑扩充规则,较少考虑到牌手与牌手之间的相互配合。结合数据库智能化的思想,构建了增加竞争与合作标识的叫牌数据库。使其能够描述桥牌比赛中存在的合作竞争关系。此外,通过引入模糊数学中模糊集合的概念,使得牌局中的同一叫品信息可以表达多重含义。给予桥牌AI在计算上的支持,使得AI可以更多地将重心放在算法部分,部分忽略叫牌中复杂的叫牌体系。

关键词：计算机博弈桥牌数据库合作与竞争模糊数学

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Q学习实现亚马逊棋评估函数自调参

引用

重庆理工大学学报（自然科学） 2022年第12期36卷 136-141页

作者：邱虹坤王浩宇王亚杰沈阳航空航天大学计算机学院沈阳110136 沈阳航空航天大学工程训练中心沈阳110136

在亚马逊棋评估函数模型中进行参数调试,主要由人工依靠经验反复实验来实现,效率较低且无法保证精度。针对人工调参效率低下、精确度不足的问题,可借助机器学习的方法来弥补。采用强化学习中Q学习的思路,构造一种具有自学习能力的网络结... 详细信息

在亚马逊棋评估函数模型中进行参数调试,主要由人工依靠经验反复实验来实现,效率较低且无法保证精度。针对人工调参效率低下、精确度不足的问题,可借助机器学习的方法来弥补。采用强化学习中Q学习的思路,构造一种具有自学习能力的网络结构,利用计算机自身反复模拟对局与迭代,实现评估函数调参工作的自动化。实验结果表明:当训练达10000次时,模型各结点Q值会趋于收敛,说明此时程序可以做出稳定合理的调参操作;在博弈实战中,模型调参后的程序也表现出了较强的棋力。

关键词：计算机博弈亚马逊棋强化学习 Q学习评估函数

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

基于数据库和经验分析的桥牌混合策略打牌模型

引用

重庆理工大学学报（自然科学） 2021年第12期35卷 134-139,154页

作者：邱虹坤郑晓东王亚杰沈阳航空航天大学计算机学院沈阳110136 沈阳航空航天大学工程训练中心沈阳110136

基于桥牌博弈模型现状,分别从3个方面提出模型优化思路,得到最终的混合策略打牌模型:对于叫牌,采用更为适合计算机理解的精确叫牌体系,运用数据库对于精确叫牌体系造成的代码冗余问题,进行二次优化;对于打牌,在开局阶段根据经验制定不... 详细信息

基于桥牌博弈模型现状,分别从3个方面提出模型优化思路,得到最终的混合策略打牌模型:对于叫牌,采用更为适合计算机理解的精确叫牌体系,运用数据库对于精确叫牌体系造成的代码冗余问题,进行二次优化;对于打牌,在开局阶段根据经验制定不同打牌方位的出牌策略,采用经验分析与明手求解策略混合模式,通过数学分析,发现2种模式交换的分割线,从而实现混合模式打牌策略;对于评估函数,通过提高细化度,使其更精确地描述牌局现状。最后,通过对比实验,验证了优化后的桥牌打牌模型的综合能力有所提升。

关键词：桥牌非完备信息博弈叫牌打牌评估

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：